This plugin draws a two-dimensional plot between TPR and FPR at different thresholds. Higher AUC score indicates a better model. By Decrease the threshold value, get more TP. If the predicted label is greater than the threshold, then it is classified as a positive prediction; otherwise it is a negative prediction.
| input | Dataset CSV file. Output results are shown in the 'Evaluation' tab. Default output_result.csv is used as input for drawing of PR curve. |
| target_variable | Target label in csv file. |
| output_variable | Predicted label in csv file. For multiclassification give the variables as(y'__0,y'__1, ... , y'__n) |
| width | Plot width, to be drawn in inches. |
| height | Plot height, to be drawn in inches. |